omnivoice-amharic
Lab-et
Speech recognition model for Amharic, English. Fine-tuned from k2-fsa/OmniVoice.
Lab-et
Builder · part of the Ethiopian AI ecosystem
Amharic Corpus — Clean Speech 146,359 clips · 408.5 hours · Amharic (አማርኛ) · 16 kHz · synthetic, studio-quality channel A large-scale synthetic Amharic speech corpus generated entirely by OmniVoice, t
1
models
1
datasets
0
repos
0
spaces
768
downloads/mo
Footprint shape — each axis normalized against the ecosystem-wide maximum. A fuller polygon means this org publishes across every channel.
